R Markdown编织报‘object not found’错误：单独运行代码正常的排查

问题详情 #

编织R Markdown报告时触发以下错误：

Error in ggplot(data = bio1530_sci1420_summary_stats.xlsx) :
object 'bio1530_sci1420_summary_stats.xlsx' not found
Calls:  ... withVisible -> eval_with_user_handlers -> eval -> eval -> ggplot
Execution halted

代码结构如下，单独运行ggplot代码块可正常生成散点图，但编织整体报告失败：

---
title: "NGRMarkdown"
author: "Rob McCandless"
date: "`r Sys.Date()`"
output: word_document
---

knitr::opts_chunk$set(echo = TRUE)
library(ggplot2)
library(ggrepel)
library(tidyverse)
library(here)
read_csv("bio1530_sci1420_summary_stats.xlsx")

#ScatterPlot of mean course grade v. mean normalized gain on 1420 and 1530 data with regression lines and error bars
ggplot(data=bio1530_sci1420_summary_stats.xlsx)+
  geom_errorbar(aes(x=Course_grade, y=Norm_gain, ymin=Norm_gain-CI, ymax=Norm_gain+CI), color="black", width=0.2, position=position_dodge2(10.0))+
  geom_point(mapping=aes(x=Course_grade, y=Norm_gain, shape=Course, color=Course),size=3)+
  geom_smooth(method=lm, se=FALSE, col='black', size=1, mapping=aes(x=Course_grade, y=Norm_gain, linetype=Course))+
  geom_label_repel(aes(Course_grade, y=Norm_gain, label = Alpha), box.padding = 0.3, point.padding = 0.7, segment.color = 'grey50')+  #added point labels A-J 
  ylab('Mean Normalized Gain (all instructor sections)')+
  xlab('Mean Course Grade (all instructor sections)')+
  scale_fill_discrete(labels=c("Bio 1530", "Sci 1420"))+
  labs(title="Normalized Gain v. Course Grade by Course & Instructor", subtitle="Mean and 95% CI of all sections per instructor (A-J)")+
  theme(plot.title=element_text(hjust=0.5))+
  theme(plot.subtitle=element_text(hjust=0.5))+
  annotate("text", x=73.0, y=0.09, label="R2 = 0.68, p = 0.044")+
  annotate("text", x=78.5, y=0.22, label="R2 = 0.46, p = 0.095")

已确认交互环境下工作目录为/Users/robmccandless/Library/Mobile Documents/com~apple~CloudDocs/R Projects/Normalized_Gain_Data，且RMD文件与数据文件在同一目录下。

错误原因 #

• 数据未赋值给对象：read_csv("bio1530_sci1420_summary_stats.xlsx")仅读取数据但未将其存储为变量，后续ggplot直接用文件名作为数据对象，自然找不到。
• 文件格式不匹配：read_csv用于读取CSV文件，无法处理XLSX格式，这会导致数据读取失败（交互环境可能因缓存或其他巧合暂时可用，但编织时会暴露问题）。
• 编织工作目录差异：R Markdown编织时的工作目录可能与交互环境不同，即使文件在同一目录，也可能因路径解析问题找不到文件。

解决方案 #

步骤1：修正数据读取逻辑 #

• 安装并加载readxl包（专门用于读取Excel文件）
• 将读取的数据赋值给一个变量，比如df
• 结合here包确保路径解析正确（避免工作目录差异问题）

步骤2：修改ggplot代码使用正确的对象 #

将ggplot(data=bio1530_sci1420_summary_stats.xlsx)改为使用赋值后的变量df

修正后的完整代码 #

---
title: "NGRMarkdown"
author: "Rob McCandless"
date: "`r Sys.Date()`"
output: word_document
---

knitr::opts_chunk$set(echo = TRUE)
# 安装readxl（首次运行需执行）
# install.packages("readxl")
library(ggplot2)
library(ggrepel)
library(tidyverse)
library(here)
library(readxl)

# 读取数据并赋值给对象，用here包指定路径
df <- read_excel(here("bio1530_sci1420_summary_stats.xlsx"))

#ScatterPlot of mean course grade v. mean normalized gain on 1420 and 1530 data with regression lines and error bars
ggplot(data=df)+
  geom_errorbar(aes(x=Course_grade, y=Norm_gain, ymin=Norm_gain-CI, ymax=Norm_gain+CI), color="black", width=0.2, position=position_dodge2(10.0))+
  geom_point(mapping=aes(x=Course_grade, y=Norm_gain, shape=Course, color=Course),size=3)+
  geom_smooth(method=lm, se=FALSE, col='black', size=1, mapping=aes(x=Course_grade, y=Norm_gain, linetype=Course))+
  geom_label_repel(aes(Course_grade, y=Norm_gain, label = Alpha), box.padding = 0.3, point.padding = 0.7, segment.color = 'grey50')+  #added point labels A-J 
  ylab('Mean Normalized Gain (all instructor sections)')+
  xlab('Mean Course Grade (all instructor sections)')+
  scale_fill_discrete(labels=c("Bio 1530", "Sci 1420"))+
  labs(title="Normalized Gain v. Course Grade by Course & Instructor", subtitle="Mean and 95% CI of all sections per instructor (A-J)")+
  theme(plot.title=element_text(hjust=0.5))+
  theme(plot.subtitle=element_text(hjust=0.5))+
  annotate("text", x=73.0, y=0.09, label="R2 = 0.68, p = 0.044")+
  annotate("text", x=78.5, y=0.22, label="R2 = 0.46, p = 0.095")

额外验证（可选） #

若仍有问题，可在数据读取代码块中添加以下代码，查看编织时的工作目录及文件列表，确认数据文件是否存在：

print(getwd())
list.files()

内容的提问来源于stack exchange，提问作者Rob McCandless